Ordinary Differential Equation class-1 lecture notes

Mohin Uddin
Department of Mathematics
Bangladesh University of Engineering and Technology, BUET

Previous question analysis for BUET admission test

1. 1st order Linear Differential Equation
[October-18]

2. Equations reducible to 1st order linear form: Bernoulli Equation

[April-18]

3. Higher Order Linear Differential Equation

[October-24, April-22, April-19, October-18]

4. Higher Order Linear Differential Equation with Variable Coefficient

[April-16, October-15, October-17]

5. Method of Variation of parameters

[April-24]

Definition of Ordinary Differential Equation:

An equation involving ordinary derivatives of one dependent variable with respect to
one independent variable is called an ordinary differential equation.
Examples:
dy
√
(i) x dx + y = y ln x
d3 y 2
(ii) dx3
− 6 dx2 + 11 dx − 6y = 0
d y dy

Order of an Ordinary Differential Equation:

The order of the highest ordered derivative involved in a differential equation is called
the order of the differential equation.
Examples:
d2 y dy
(i) dx2
+ 2 dx +y =0

1
Here, the highest derivative is of second order. So, this equation is said to be a
second-order O.D.E.

( )
d2 y 3
(ii) dn2
+ a2 y = 0

Eq(ii) has a maximum derivative of 2. Hence, it is said to be a second-order O.D.E.

Degree of an O.D.E.
The degree of the highest ordered derivative involved in a differential equation is
called the degree of the differential equation, when it has been made free from
fractions and negative.
Examples:

( )2
d2 x dx
(i) dt2
+2 dt
+ 7x = 0
d2 x
The highest derivative dt2
is raised to a power 1. So the differential equation is of
degree 1.

√
d2 y dy
(ii) dx2
+ x dx + y = cos x

This can be written as

d2 y dy
• dx2
+ x dx + y = cos2 x

Hence the degree of the O.D.E. is 1.

Solution of a Differential Equation:

If any differential equation is satisfied by any defined and differentiable function, then
the function is called a solution of the differential equation.

Explicit Solution:
If the function y = f (x) is the solution of a differential equation of variables x and y,
then the solution is called an explicit solution.

Implicit Solution:
If the function F (x, y) = 0 is the solution of a differential equation of variables x and
y, then it is called an implicit solution.
Example:
x2 + y 2 = c is an implicit solution of the equation 2x dx + 2y dy = 0.

2
General or Complete Solution:
If the solution of any nth order differential equation contains n arbitrary constants,
then the solution is called the general solution of the differential equation.

Particular Solution:
The solution obtained for the definite values of the arbitrary constants of the general
solution is called the particular solution of the differential equation.

Singular Solution:
If any other solution except the general or particular solution of a differential equation
exists, then it is called a singular solution of the differential equation.

First Order and First Degree Differential Equation:

Type–1: Variable Separable Equations
Definition: If the differential equation M dx + N dy = 0 can be expressed in the form
f1 (x) dx = f2 (y) dy,, then it is called a variable separable equation. Variables can
be separated directly or by method of substitution.

(i) Direct Variable Separable Equation :

( )
dy
Problem-1: Solve: ln dx
= ax + by

Solution:
Given,
( )
dy
ln = ax + by
dx
dy
⇒ = eax+by
dx
dy
⇒ = eax · eby
dx
⇒e−by dy = eax dx
∫ ∫
⇒ eax dx − e−by dy = 0 [Integrating]
eax e−by
⇒ + +C =0
a b
⇒beax + ae−by + A = 0 [A = abc]

Ans.

3
Problem-2: Solve: y dx − x dy = xy dx

Solution:
Given,

y dx − x dy = xy dx
⇒y(1 − x) dx = x dy
1−x dy
⇒ dx =
x y
( )
1 dy
⇒ − 1 dx =
x y
∫ ( ) ∫
1 dy
⇒ − 1 dx = [Integrating]
x y
⇒ ln x − x = ln y + c
( )
x
⇒ ln =x+c
y

Ans.

Another Solution:

y dx − x dy = xy dx
y dx − x dy
⇒ = dx
xy
dx dy
⇒ − = dx
x y
⇒ ln x − ln y = x + c [Integrating]
( )
x
⇒ ln =x+c
y

Ans.

(ii)Separable of variables by substitution:

( )
Problem-3: Solve: sin−1 dx
dy
=x+y
[NTRCA(College: 16,12,7,5th; School:17th)]
Solution:

4
Given,
( )
−1 dy
sin =x+y
dx
dy
⇒ = sin(x + y)
dx
dv
⇒ − 1 = sin v
dx Let, x + y = v
dv dy dv
⇒ = 1 + sin v ⇒1+ =
dx dx dx
dv dy dv
⇒ = dx ⇒ = −1
1 + sin v dx dx
1 − sin v
⇒ dv = dx
cos2 v
⇒(sec2 v − sec v tan v)dv = dx
⇒ tan v − sec v = x + c (Integrating)
⇒ tan(x + y) − sec(x + y) = x + c

Ans.

Problem-4: Solve: y(1 + xy) dx + x(1 − xy) dy = 0

[NTRCA(College: 17,8th; School:13th)]

Solution:
Given,

y(1 + xy) dx + x(1 − xy) dy = 0

⇒ y dx + xy 2 dx + x dy − x2 y dy = 0
⇒ y dx + x dy + xy(y dx − x dy) = 0
⇒ d(xy) + xy(y dx − x dy) = 0
d(xy) y dx − x dy
⇒ + =0
(xy)2 xy
d(xy) 1 1
⇒ 2
+ dx − dy = 0
(xy) x y
1
⇒ − + ln x − ln y = c [By integrating]
xy
1 x
⇒ − + ln = c
xy y
x 1
∴ ln − =c
y xy
Which is the required solution.

Problem-5: Solve: (x + y) dy + (x − y) dx = 0

5
Solution:
Given,

(x + y) dy + (x − y) dx = 0
⇒ x dy + y dy + x dx − y dx = 0
1
⇒ x dy − y dx + d(x2 + y 2 ) = 0
2
x dy − y dx d(x2 + y 2 )
⇒ + =0
x2 + y 2 2(x2 + y 2 )
x dy−y dx
d(x2 + y 2 )
⇒ x2
( )2 + =0
1 + xy 2(x2 + y 2 )
( )
y
d x d(x2 + y 2 )
⇒ ( )2 + =0
1+ y 2(x2 + y 2 )
x
( )
−1 y 1
⇒ tan + ln(x2 + y 2 ) = c
x 2
Which is the required solution.

Note -1: ( ) ( )
d y x dy − y y x dy − y dx
= dx 2 , d =
dx x x x x2

Note-2:
1 d 2 2 1 dy dy x dx + y dy 1
(x +y ) = (2x+2y ) = x+y = ⇒ d(x2 +y 2 ) = x dx+y dy
2 dx 2 dx dx dx 2

Type–2: Homogeneous Differential Equation

Definition: The differential equation M (x, y) dx + N (x, y) dy = 0 is called an n-
dimensional homogeneous differential equation if and only if
( ) ( )
n y n y
M (x, y) = x M1 , N (x, y) = x N1
x x
or, ( ) ( )
n x n x
M (x, y) = y M2 , N (x, y) = y N2
y y
Shortcut for Homogeneous Differential Equation:
If M (x, y) dx + N (x, y) dy = 0 is a homogeneous differential equation, then the trans-
formation y = vx will separate the variables.

6
 ( )
dy y y
Problem 1: Solve: dx
= x
+ tan x
[NTRCA(College: 14,10th; School:15th)]
Solution:

Given,
( )
dy y y
= + tan
dx x x
dv vx
⇒v+x = + tan(v)
dx x
dv
⇒v+x = v + tan(v)
dx
dv Let, y = vx
⇒x = tan v dy dv
dx ⇒ =v+x
dx dx dx
⇒ cot v dv =
x
⇒ ln(sin v) = ln x + ln c [Integrating]
⇒ sin v = cx
( )
y
⇒ sin = cx,
x
Which is the required solution.

Problem-2: Solve: (x2 + y 2 ) dx = 2xy dy

[NTRCA(College: 12th, School: 14, 18th)]
Solution:
Given,

(x2 + y 2 ) dx = 2xy dy
dy (x2 + y 2 )
⇒ =
dx 2xy
dv x2 + v 2 x2 1 + v2
⇒v+x = =
dx 2x · vx 2v
2
dv 1+v
⇒x = −v
dx 2v
dv 1 − v2
⇒x =
dx 2v Let, y = vx
2v dx
⇒ dv = dy dv
1−v 2 x ⇒ =v+x
2v dx dx dx
⇒ 2 dv + =0
v −1 x
⇒ ln(v 2 − 1) + ln x = ln c [Integrating]
⇒ x(v 2 − 1) = c
( )
y2
⇒x 2 −1 =c
x
( )
⇒ y 2 − x2 = cx
⇒ y 2 − x2 = cx

7
Which is the required solution.

Problem-3: Solve: (x + y) dy + (x − y) dx = 0
[NTRCA(College: 4th)]
Solution:
Given,

(x + y) dy + (x − y) dx = 0
dy x + y
⇒ + =0
dx x − y
dv x − vx
⇒v+x + =0
dx x + vx
dv 1 − v
⇒v+x + =0
dx 1 + v
dv v(1 + v) + 1 − v
⇒x + =0
dx 1+v
dv 1 + v 2 Let, y = vx
⇒x + =0
dx 1+v dy dv
1+v dx ⇒ =v+x
⇒ v dv + =0 dx dx
1+v 2 x
1 v dx
⇒ dv + dv + =0
1 + v2 1 + v2 x
1
⇒ tan−1 v + ln(1 + v 2 ) + ln x = c
2 √ 
( ) ( )2
y y
⇒ tan−1 + ln  + 1 · x = c
x x
( ) √
−1 y
⇒ tan + ln (x2 + y 2 ) = c
x
Which is the required solution.

Type–3: Exact Differential Equation

Definition: The differential equation M dx + N dy = 0 will be an exact differential
equation if M dx + N dy = du, where u = u(x, y)

Example:

y dx + x dy = 0
⇒y dx + x dy = d(xy) = 0

Another example of exact differential equation:

ex cos y dy + ex sin y dx = 0
⇒d(ex sin y) = 0

8
Necessary and sufficient condition for the exactness of the differential equation M dx+
N dy = 0 is
∂M ∂N
=
∂y ∂x
Where,
∂u
M=
∂x
and
∂u
N=
∂y

Solved Examples from Exact Differential Equation:

Problem-1: Solve: (2x + 3y + 6) dy = (6x − 2y − 7) dx

Solution:
Given,

(2x + 3y + 6) dy = (6x − 2y − 7) dx
⇒(6x − 2y − 7) dx + (−2x − 3y + 6) dy = 0

Here,
M = 6x − 2y − 7, N = −2x − 3y + 6

∂M ∂N
= −2, = −2
∂y ∂x
∂M ∂N
Since, ∂y
= ∂x
, the given differential equation is exact.

Required solution is:

∫ ∫
M dx + N dy = c
y constant x free
∫ ∫
⇒ (6x − 2y − 7) dx + (−3y + 6) dy = c
y constant
2 ∫
x
⇒6 · − 2xy − 7x + (−3y + 6) dy = c
2 ∫
⇒3x2 − 2xy − 7x + (−3y + 6) dy = c
3
⇒3x2 − 2xy − y 2 + 6y = c
2
⇒6x − 4xy − 3y 2 − 14x + 12y = c1
2

Ans.

9
Problem-2: Solve: (x3 + 3xy 2 ) dx + (y 3 + 3x2 y) dy = 0
[NTRCA(School – 10th)]
Solution:
Here,
M = x3 + 3xy 2 , N = y 3 + 3x2 y

∂M ∂N
= 6xy, = 6xy
∂y ∂x
∂M ∂N
Since ∂y
= ∂x
, the given differential equation is exact.

Required solution is:

∫ ∫
M dx + N dy = c
y constant x free
∫ ∫
⇒ 3 2
(x + 3xy ) dx + y 3 dy = c
y constant
x4 3 y4
⇒ + x2 y 2 + =c
4 2 4
⇒x4 + 6x2 y 2 + y 4 = c1

Ans.

From here important for BUET admission test

Type–4: Linear Differential Equation
The differential equation of the form
dy
+ P (x)y = Q(x)
dx
is called a first-order linear differential equation.

Another form:
dx
+ P (y)x = Q(y)
dy
If Q(x) = 0, then it will be called a first order homogeneous equation; otherwise,
non-homogeneous linear differential equation.

Integrating Factor (I.F) of linear differential equation:

If the differential is of the form:

dy
(i) dx
+ P (x)y = Q(x)

10
Then the Integrating Factor (I.F) will be:
∫
P (x) dx
I.F = e

If the differential be of the form:

dx
(ii) dy
+ P (y)x = Q(y)

Then the Integrating Factor (I.F) will be:

∫
Q(y) dy
I.F = e

Solved Examples from Linear O.D.E:

dy
Problem-1: Solve: dx + 2xy = x
[NTRCA(College-9th)]

Solution:

Here, P (x) = 2x, Q(x) = x

∫
2x dx 2
Now, I.F = e = ex

Required solution is:

∫ ∫ ∫
P (x) dx P (x) dx
ye = Q(x)e dx
∫
2 2
⇒ yex = xex dx + c
1∫ z
= e dz + c Let, x2 = z, ⇒ 2xdx = dz
2
1 z 1
= e +c ⇒ xdx = dz
2 2
1 x2
= e +c
2
1
+ ce−x
2
⇒y=
2
Which is the required solution.
dy
Problem-2: Solve: (1 + x2 ) dx + y = tan−1 x
[NTRCA(College-6th)]
Solution:

Given,
dy
(1 + x2 ) + y = tan−1 x
dx

11
Which can be written as,
dy y tan−1 x
⇒ + =
dx 1 + x2 1 + x2
1 tan−1 x
Here, P (x) = 1+x2
, Q(x) = 1+x2
∫ ∫ 1
dx −1
Now, I.F. = e P (x) dx = e 1+x2 = etan x

Required solution is:

∫ ∫ ∫
P (x) dx P (x) dx
ye = Q(x)e dx
∫
−1 tan−1 x tan−1 x
⇒ yetan x
= e dx
∫ 1+x
2 Let, u = tan−1 x
= ueu du 1
∫
⇒ du = dx
1 + x2
∫ ∫ ∫ ( )
= ue − du ∫
u
eu du
N ote : u v dx = u v dx − v dx dx
= ueu − eu + c dx
−1 −1 −1
⇒ yetan x
= tan−1 x · etan x
− etan x
+c
−1 − tan−1 x
⇒ y = tan x−1+c·e
Which is the required solution.

Problem-3: Solve: (1 − x2 ) dx
dy
− xy = 1
[NTRCA(College-14th)]

Solution:

Given,
dy
(1 − x2 ) − xy = 1
dx
Which can be written as,
dy x 1
⇒ − · y =
dx 1 − x2 1 − x2
−x 1
Here, P (x) = 1−x2
, Q(x) = 1−x2
∫ ∫ −x √ √
dx 1 2
Now, I.F. = e P (x) dx
=e 1−x2 = e 2 ln(1−x ) = eln( 1−x2 )
= 1 − x2
Required solution is:
∫ ∫ ∫
P (x) dx
ye Q(x)e P (x) dx dx
=
√ ∫ √
1 − x2
⇒ y 1 − x2 = dx ∫
∫
1 − x2 1
1 Here, √ dx = sin−1 x
= √ dx 1−x 2
1 − x2
= sin−1 x + c
√
⇒ y 1 − x2 = sin−1 x + c

12
Which is the required solution.
dy
Problem-4: Solve: cos2 x + y = tan x
dx

Solution:

Given,
dy
cos2 x + y = tan x
dx
Which can be written as,
dy 1 1
+ 2
·y = · tan x
dx cos x cos2 x
dy
⇒
+ sec2 x · y = sec2 x · tan x
dx
Which is a linear differential equation of order one.

Here, P (x) = sec2 x, Q(x) = sec2 x · tan x

∫ ∫
P (x) dx sec2 x dx
Now, I.F. = e =e = etan x
Required solution is:
∫ ∫ ∫
P (x) dx P (x) dx
ye = Q(x)e dx
∫
⇒ yetan x = sec2 x · tan x · etan x dx
∫ Let u = tan x
= tan x · e tan x
d(tan x) ⇒ du = sec2 x dx
∫ ∫ ∫
= tan x · e tan x
− e tan x
d(tan x) Then, ueu du = ueu − eu du

= tan x · etan x − etan x + c

⇒ y = tan x − 1 + c · e− tan x

Which is the required solution.

Equations Reducible to First Order Linear Equation:

Bernoulli’s Equation:
dy
The differential equation of the form dx
+ P y = Q(x)y n is called a Bernoulli equa-
tion.

Strategy to solve Bernoulli Equation:

The general form of the Bernoulli differential equation is:
dy
+ P (x)y = Q(x)y n
dx

13
Now, divide both sides by y n , which gives:
dy
y −n + P (x)y 1−n = Q(x)
dx
Now let:
dz dy
z = y 1−n ⇒ = (1 − n)y −n
dx dx
Which eventually reduced to a linear differential equation of some function of y.
Then, we can easily solve the reduced linear differential equation by known methods.

Solved Examples from Bernoulli’s Equation:

dy
Problem-1: Solve dx + y = ex y 2
[NRTCA (College – 12th, School – 6th)]

Solution:

Given,
dy
+ y = ex y 2
dx
Which can be written as,

1 dy 1
⇒ + = ex
y 2 dx y
( )
d 1 1
⇒− − = −ex
dx y y
( )
d 1 1
⇒ − = −ex
dx y y

This is a linear differential equation in y1 .

P (x) = −1, Q(x) = −ex

Now, ∫ ∫
−1 dx
I.F. = e P (x) dx
=e = e−x
Required solution is:
1 −x ∫
· e = −ex · e−x dx
y ∫ 1
Let, u =
= −1 dx = −x + c y
1 du
⇒ = ex (−x + c) = (c − x)ex ⇒ − u = −ex
y dx
1 This is a linear ODE in u
⇒y=
(c − x)ex

14
Or, equivalently:
(c − x)y = e−x Ans.
dy 3
Problem-2: Solve: x dx + y = (xy) , 2 y(1) = 4
[BUET–18]

Solution:

Given differential equation:

dy 3 dy 3 3
x + y = (xy) 2 ⇒ x + y = x 2 y 2
dx dx
Divide both sides by x:
dy 1 1 3
+ y = x2 y 2
dx x
Multiply both sides by y − 2 :
3

dy 1
y− 2 + y− 2 · = x 2
3 1 1

dx x
Which gives:
( ) 1
d −1 1 x2
y− 2 = −
1
(y 2 ) + −
dx 2x 2
Which is a linear differential equation in y − 2 of order one.
1

Integrating Factor:
∫ −1/2 1
− 2x
= e− 2 ln x = eln x
1 1
I.F. = e dx
=√
x
General Solution is:
∫
− 121 x1/2 1
y ·√ = − · √ dx
x 2 x
Let: u = y − 2
1
∫
1 x
= − dx = − + c du 1 x1/2
2 2 Then: − u=−
1 x dx 2x 2
⇒ √ =− +c
xy 2
Using Initial Condition: x = 1, y = 4
1 1 1 1
√ =− +c⇒ =− +c⇒c=1
1·4 2 2 2
Particular Solution:
1 x
√ =− +1 Ans.
xy 2

Problem-3: Solve: (1 − x2 ) dx
dy
+ xy = xy 2 ; (1 > x)
[NTRCA(College: 10th)]

15
Solution:
Given,
dy
(1 − x2 )
+ xy = xy 2
dx
1 dy x 1 x
⇒ 2 + · =
y dx 1 − x y2 1 − x2
1 dy x 1 x
⇒− − · =−
y dx 1 − x y
2 2 1 − x2
( )
d 1 x 1 −x
⇒ − · =
dx y 1−x y
2 1 − x2
( )
1
This is a linear differential equation of y
of order one.
−x −x
Here, P (x) = , Q(x) =
1 − x2 1 − x2
Integrating Factor:
∫ ∫ −x
P dx dx
I.F = e =e 1−x2

1
∫ −2x
dx
=e 2 1−x2

1 2 √
= e 2 ln(1−x ) = 1 − x2

General solution is:

1 ∫ P dx ∫ ∫
·e = Q · e P dx dx
y
∫
1√ −x √
⇒ 1 − x2 = · 1 − x2 dx
y 1 − x2
√ ∫
1 − x2 x 1 ∫ −2x
⇒ =− √ dx = √ dx
y 1 − x2 2 1 − x2
√
1 − x2 1 √
⇒ = · 2 1 − x2 + c
y 2
√ √
⇒ 1 − x = y 1 − x2 + cy
2
√
⇒(1 − y) 1 − x2 = cy when x < 1

Similarly, √
(1 − y) x2 − 1 = cy when x > 1

16
Modeling with First Order Differential Equations:

Problem-3: If the population growth rate is proportional to the total population at

any fixed time, and if it doubles in 50 years, how long will it take to become 3 times
as much?
[NTRCA(School: 15,14,13th)]
Solution:

Let the population of a city be x at any fixed time t. Since the rate of increase is
proportional to the quantity,
dx dx
∝x⇒ = kx (1)
dt dt
Separating variables:
dx
= k dt
x
Let at t = 0, the quantity is x0 , and after t = 50, the quantity becomes 2x0 . Then
integrating both sides,

∫ 2x0
dx ∫ 50
= k dt
x0 x 0

[ln x]2x0 50
x0 = k [t]0
⇒ ln 2x0 − ln x0 = k · 50
( )
2x0
⇒ ln = 50k
x0
⇒ ln 2 = 50k

Therefore,
ln 2
k=
50
Now,
At t = 0, let x = x0 and at t = t, x = 3x0 .

We now determine how long it takes for the quantity to triple.

∫ 3x0
dx ∫ t
= k dt
x0 x 0

k [t]t0 = [ln x]3x

x0
0

kt = ln(3x0 ) − ln(x0 )
( )
3x0
kt = ln = ln 3
x0
ln 2
From earlier, we had k = 50
, so:

17
 50kt = 50 ln 3
t · ln 2 = 50 ln 3
50 ln 3
t=
ln 2
t ≈ 79 (years)

Hence, it will take approximately 79 years.

The End

18